Preference Satisfaction versus Desire Fulfilment

faithlessgod · by **faithlessgod** on 2008-11-26T10:29:00

Ryan asked me a question in the Desire Utilitarianism thread to show him a difference between DU and PU. This is my take on this and I would welcome feedback from any PU advocates here. This is not a definitive argument but certainly an illustrative one of the differences. I specifically covered this in a blog post The Problem with Utilitarianism. The below is a partial cut and paste from this but I humbly recommend you read the whole post

Preference Satisfaction
The simplest statement of this is that everyone has preferences, preference satisfaction is the utility to maximize and that utilitarianism is the technique to maximize this utility. What this means is that utilitarian maximizing of preferences satisfaction is "good" and its opposite is "bad", however there are some problems with this approach, three are highlighted here and, it is argued, that Desire Utilitarianism resolves.

Differential Preferences
A bias is introduced due the fact that different agents, due to differing circumstances in terms of health, nutrition, gender, sexual preference, ethnicity, age, socioeconomic status, influence and expectation and so on, have quantitatively, as well as qualitatively, different preference sets. Stated in a simplistic fashion poorer people have fewer preferences that need to be satisfied (unless they are not a realist about their situation) and that richer people can afford to and can have more preferences to satisfy. Given the Pareto distribution of rich and poor, there are fewer rich agents with more preferences and a larger amount of poor agents with fewer preferences. There are then two ways to maximize preferences: on a per agent basis one would choose to maximize the preferences of as many agents as possible which would bias satisfying the poorer agents over the richer agents. Another way to maximize preferences would be on a per preference basis which would tend to maximize the preferences of richer agents - having proportionately more preferences over those poorer agents with the fewest preferences. Of course, depending on the mix of "rich" and "poor" and the average amount of preferences of these two groups, what preferences are satisfied or not will also change and this leads to the next point.

Demographic Bias
If one takes the challenge to maximise preferences on a local or societal basis - on the basis that the only preferences to be satisfied must be those within the scope of the society that is capable of doing so, then this becomes sensitive to the demographic distribution of preferences. Whilst religion and ethnicity are only two of the causal factors that contribute to such demographics, they are very illustrative of the problem. Change the demographic mix and what preferences to satisfy change. This looks like cultural relativism of a sort but I am not trying to make that specific point here. Since the problem still occurs if one takes a humanist spin and so includes all the preferences of all the people on the planet at one time. It is still the case that the demographic mix of the planet changes through time and so the preference to satisfy can change. This analysis still applies even if one normalizes out resources and other non-demographic factors.

One answer, obliquely related to both points made here, is to exclude external preferences from what is to be maximized:

External Preferences
Many if not most people have preferences that can only be satisfied based on their effects on other's preferences. The recognition of this and the exclusion of external preferences is an ad hoc means to dealing with this problem . This is of a majority biased against a minority having external preferences that whose satisfaction ensures this bias. Examples includes social, economic, religious, political, sexual and gender discrimination, apartheid, bigotry and so on. It is as ad hoc method and it also removes any preferences that people might have to help fulfill other's preferences such as charity and so on.

Conclusion - Desire Utilitariansim resolves these biases
[Desire Utilitarianism]...deals with these three objections and problems in, let us call it, naive Preference Satisfaction Utilitarianism.I am not saying that sophisticated versions cannot deal with this, indeed Desire Utilitarianism could considered as such. I am saying that to deal with this something like Desire Utilitarianism, if presenting an alternative, needs to be done.

DanielLC · by **DanielLC** on 2009-01-06T19:39:00

I still don't understand the difference. I made a proof that the two forms are the same: http://scratchpad.wikia.com/wiki/Talk:Frequently_Asked_Questions_about_Desire_Utilitarianism. Point out the flaw in the proof. I think that would be the best way to explain the difference to me.

faithlessgod · by **faithlessgod** on 2009-01-07T00:13:00

Daniel

First of all the above note is a little old na dI think I could do better now. Nonetheless your challenge does not any of the points I addressed in the first post.

Lets look at your challenge. I have to rewrite this the way I understand DU partly because there is some oddness in the way you state these. It could be in my rewrite that I have adjusted things too much from what you argued for but then this will only help in isolating these differences.

OK here goes

1. A desire is what a person will strive to achieve. (Axiom)

1A) A person seeks to fulfil their desires.
1B) A desire is an attitude to make or keep something true
The definition of "desire" is not dependent on "person".

2. A good desire is a desire that will help achieve other desires. (Axiom)

2A) A good desire is such as to fulfil or tend to fulfil other desires.

3. What you should do is what someone with a good desire would do. (Axiom)

3A) The right act is the act that a person with good desires would perform

4. What you should do is what someone who has a desire that will help achieve other desires would do. (Substituting good desire in 3 with its definition in 2)

4A)The right act is the act that a person with desires that tend to fulfil other desires would perform (substituting "good desire" in 3A with its definition in 2A)

5. What you should do is what someone who will strive to help achieve other desires would do. (Substituting desire in 3 with its definition in 1)

5A)The right act is the act that a person with good attitudes to make or keep something true would perform (Substituting desire in 3 with its definition in 1B)
Aha! "good" was dropped in 5! And the definition of desire in 1 is mistaken

6. What you should do is what someone who will strive to help achieve desires would do. (The desire that "other" excludes is no longer mentioned, so "other" can be taken out)

7. What you should do is help achieve desires. (Hard to explain, but I think you can understand that step)

5 and 1 needs to be corrected??

DanielLC · by **DanielLC** on 2009-01-07T04:53:00

Step 5 was supposed to be "Substituting desire in 4 with its definition in 1". I have now fixed that mistake on the website.

This would change 5A to: The right act is the act that a person with attitudes to make or keep something true that tend to fulfill other desires would perform

I'm confused about 'The definition of "desire" is not dependent on "person"'. Certainly, there can't be a desire without a person. You defined a desire as an attitude, which I think we can all agree exists solely as an attribute of a person.

I may have figured out the problem. I was implying having a desire that tends to achieve something is the same as desiring to achieve it. A desire that tends to achieve something is a desire such that, if someone had that desire, they'd be more likely to achieve that. If I understand this correctly, that is not necessarily someone in your situation.

Of course, if that is right, there is a pretty obvious problem. Let's take penicillin, for example. The desire to take penicillin when you get sick would be a bad desire, as a lot of people get allergic reactions to penicillin. The desire to avoid penicillin would be a bad desire, as a lot of people would get sick from it. You'd be stuck with the desire to take it when your sick only if you aren't allergic. But you're either always allergic or never allergic. Does this mean that you should train people to keep the condition on the end in fear that the people who are allergic decide to always avoid it, which is a bad desire?

Also, while you're at it, can you answer the last two questions on the main article?

faithlessgod · by **faithlessgod** on 2009-01-07T14:50:00

DanielLC wrote:Step 5 was supposed to be "Substituting desire in 4 with its definition in 1". I have now fixed that mistake on the website.
This would change 5A to: The right act is the act that a person with attitudes to make or keep something true that tend to fulfill other desires would perform.

Read it. Thanks. I think what you are trying to do is to show that desire utilitarianism collapses into desire fulfilment act utilitarianism - since the latter is explicitly just a version of preference satisfaction? That the difference is only a surface feature but they are really share the same deep structure? I will revisit this argument in a separate post.

DanielLC wrote:I'm confused about 'The definition of "desire" is not dependent on "person"'. Certainly, there can't be a desire without a person. You defined a desire as an attitude, which I think we can all agree exists solely as an attribute of a person.

We can create intelligent agents with equivalent attitudes. That is they might only naturally occur in sentient beings but we can create artificial alternates. Nonetheless axiom (1) seems to confuse definitions of desires and persons. Defining desire as an attribute of a person is incorrect, since this is not a definition of desire only a claim, correct in our view, that desires are attributes of person. A desire is defined using an attribute - the attitude - of a person

DanielLC wrote:I may have figured out the problem. I was implying having a desire that tends to achieve something is the same as desiring to achieve it. A desire that tends to achieve something is a desire such that, if someone had that desire, they'd be more likely to achieve that. If I understand this correctly, that is not necessarily someone in your situation.

"Tend" is (cough) intended as "indirect" e.g good is "such as to (directly) fulfil or (indirectly) tend to fulfil the desires of the kind in question". Alternately "direct" also corresponds to fulfilling desire-as-ends, "indirect" or "tend to" corresponds to (directly) fulfilling desires-as-means - since desire-as-means are obviously means to fulfilling desires-as-ends. (Could state this more clearly but in a hurry). Does this help?

DanielLC wrote:Of course, if that is right, there is a pretty obvious problem. Let's take penicillin, for example. The desire to take penicillin when you get sick would be a bad desire, as a lot of people get allergic reactions to penicillin. The desire to avoid penicillin would be a bad desire, as a lot of people would get sick from it. You'd be stuck with the desire to take it when your sick only if you aren't allergic. But you're either always allergic or never allergic. Does this mean that you should train people to keep the condition on the end in fear that the people who are allergic decide to always avoid it, which is a bad desire?

This looks confused or at least confuses me. Maybe with the clarification of "tend" above you could update this? I am grateful you are examining and asking these interesting question about DU.

DanielLC wrote:Also, while you're at it, can you answer the last two questions on the main article?

Sure, when I have time. They are good question IMHO.

faithlessgod · by **faithlessgod** on 2009-01-07T15:19:00

We are still trying to get the argument structure clear before deciding fully on validity, hence I am trying to update this as charitably as possible for now.

1. A desire is what a person will strive to achieve. (Axiom)

1A) A person seeks to fulfil their desires.
1B) A desire is an attitude to make or keep something true
The definition of "desire" is not dependent on "person" only on an attribute of a person - an attitude.

2. A good desire is a desire that will help achieve other desires. (Axiom)

2A) A good desire is such as to fulfil or tend to fulfil other desires.

3. What you should do is what someone with a good desire would do. (Axiom)

3A) The right act is the act that a person with good desires would perform

4. What you should do is what someone who has a desire that will help achieve other desires would do. (Substituting good desire in 3 with its definition in 2)

4A)The right act is the act that a person with desires that tend to fulfil other desires would perform (substituting "good desire" in 3A with its definition in 2A)

5. What you should do is what someone who will strive to help achieve other desires would do. (Substituting desire in 4 with its definition in 1)

5A)The right act is the act that a person with attitudes to make or keep something true that tend to fulfill other desires would perform (your suggestion)
5B)The right act is the act that a person with an attitude to make or keep tending to fulfil other desires true would perform (Substituting the first desire in 4 with its definition in 1B and substituting "something" too)

6. What you should do is what someone who will strive to help achieve desires would do. (The desire that "other" excludes is no longer mentioned, so "other" can be taken out)

6A) The right act is the act that a person with an attitude to make or keep tending to fulfil desires true would perform (The desire that "other" excludes is no longer mentioned, so "other" can be taken out)
There are dangers here in dropping the "other". You have now turned this into a "desire to fulfil desires" which can be read as a second order desire within a person. The initial framework was relations between 1st order desires. This move looks illegitimate here as it is open to such misinterpretation. Leave it for now.

7. What you should do is help achieve desires. (Hard to explain, but I think you can understand that step)

Well you will have to explain how you get to 7 from 6. Plus dropping the "other" makes this statement ambivalent - whose desires? Preference satisfaction states that
8) "the right act is the act that maximises the satisfaction of preferences", this is not the same as 7 AFAICS. Where is the maximisation? The challenge you have set yourself is to get from (3A) to (8). I do not see it yet but am interested.

DanielLC · by **DanielLC** on 2009-01-07T19:37:00

I don't quite get what constitutes a desire. Are a person's desires distinct? I get the feeling that they're not. I'm going to use the word "superdesire" to mean the collection of all desires that a person has. They have a given desire if and only if it's part of their superdesire.

The use of the word "results" in the following proof is ambiguous. As a heads up, it will always qualify the word exactly two words before it, and never the word five words before it.

1. The best superdesire is the one that results in the most desire fulfillment. (axiom)
2. The right act is the act that a person with the best superdesire would perform. (axiom)
3. A person with a superdesire that results in something is the same as a person that results in it. (axiom)
4. An act that a person that results in something would perform is the same as an act that results in it. (axiom)
5. The right act is the act that a person with the superdesire that results in the most desire fulfillment would perform. (Substituting "best superdesire" in 2 with its definition in 1)
6. The right act is the act that a person that results in the most desire fulfillment would perform. (Substituting "a person with the superdesire that results in" in 5 with its definition in 3)
7. The right act is the act that results in the most desire fulfillment. (Substituting "the act that a person that results in something would perform" in 6 with its definition in 4)

I have eliminated my old axiom one and added what are now axioms three and four. Axiom three is only correct if the best superdesire, and, by extension, good desires, are dependent upon the circumstances of the person attempting to fulfill them.

Are good desires ones that tend to fulfill other good desires, or just other desires in general. If the former, you end up with a recursive definition that is not guaranteed exactly one result. You could have a desire that is both good and bad, or a desire that is neither. If the latter, I don't see why, in the scenario in the FAQ, wanting to torture the child is a bad desire. They'd be helping fulfill 999 other desires to torture the child. Do they all only count as one desire? If so, what constitutes a different desire? If you had one sadist who wanted to torture the child, one sadist who wanted to test the torture chamber, and one hedonistic altruist who wanted the sadists to be happy, then would each of them have good desires?

I tend to suspect that things like this tend to handwave to get the intuitive answer. If you changed it to ten sadists to a child, but had 400 million people, would it be different? What if you proceeded to change the child to a rich guy, the torture to taxes that outweigh government benefits, and the sadistic pleasure to government benefits that outweigh taxes? I believe that that fairly accurately, though not precisely, portrays the American tax system.

faithlessgod · by **faithlessgod** on 2009-01-07T21:02:00

DanielLC wrote:I don't quite get what constitutes a desire. Are a person's desires distinct? I get the feeling that they're not. I'm going to use the word "superdesire" to mean the collection of all desires that a person has. They have a given desire if and only if it's part of their superdesire.

In slightly simplified terms

a desire is an attitude to make or keep something true
a belief is an attitude that something is true.

All other relevant terms can be reduced to one or other of these two attitudes, these are the only attitudes that exist
knowing, plausible, possible, probable -> belief
need, want, preference, intersts -> desire

Thats it. This is bog standard philosophical psychology.

Everyone seeks to substitute a more fulfilling state of affairs for a less fulfilling one. They do this by seeking to fulfil the more and srtonger of their desires. They act on the more and stronger of their desires, given their beliefs.

I have no idea what you mean by the construct "superdesire" it looks highly unorthodox and these axioms based on it do not make sense. There is not much to comment on until you can clarify this.

DanielLC · by **DanielLC** on 2009-01-07T22:43:00

The problem with talking about desires is that if I where to say someone has all the good desires, it wouldn't be very specific. They could still have bad desires. Even if I specified that they didn't, they would desire different things different amounts. A superdesire just means all of someone's desires, and nothing else. It completely describes everything that they desire.

When I asked about them being distinct I meant something like this:
If my desire to play on the computer and my desire to stay well-fed are exactly two desires, then they are distinct. If they can be described as one desire: to play on the computer and be well-fed, they are not.

I figure desires being distinct wouldn't make much sense, as there are an infinite number of ways I could be playing on the computer. In every one my "play on the computer" desire is fulfilled. I think the best way to think of a desire is a set of ordered pairs of every possible universe and their corresponding desire fulfillment level. If you did it this way, it would get confusing to have two desires, as each one would have a different desire-fulfillment level for each universe. You could just add them, but then there's no way to say that someone has a specific desire, because there will always be an infinite number of ways you could add other desires to describe them perfectly. Hence: superdesire, of which you have exactly one. Think of it like the superset of all your desires. Also, I know what it means. That doesn't really help, sense you clearly didn't understand it.

Axiom three essentially says that "a desire results in something" is just another way of saying "a person with that desire results in something". In order for this to be correct, when you talk about the result of the desire, you must specify, implicitly or explicitly, what specific circumstances that desire occurs under. For example: the desire to take penicillin when one gets sick will only result in allergic reactions to it if the person with that desire is allergic. This is as opposed to something like: the desire to take penicillin when one gets sick will result in everyone who's allergic to penicillin getting allergic reactions to it. In the first case, this is talking about one person having that desire. In the second, it's talking about what will tend to happen if everyone has that desire. If the first is correct, DU and DFAU are pretty much the same thing. If it's the second, I see how there's a difference, but the DU makes no sense to me, as I can't understand why whether or not you should hold a preference should depend on what would happen if other people held it.

If you understand axiom four, ignore this. You never actually asked about it, but It's a new axiom, so I probably should explain it.
Axiom four essentially says that "a person results in something" is the same as "the act they'd take results in it". For example, if you, being in my situation, would result in, say, a certain donation to charity, that would mean that if I were to do what you'd do if you were in my situation, that charity would receive the same donation.

faithlessgod · by **faithlessgod** on 2009-01-08T00:25:00

DanielLC wrote:The problem with talking about desires is that if I where to say someone has all the good desires, it wouldn't be very specific. They could still have bad desires. Even if I specified that they didn't, they would desire different things different amounts. A superdesire just means all of someone's desires, and nothing else. It completely describes everything that they desire.

And that "superdrive" is an abstraction too far. Why? When we ask, in order to find the right action, what a good person would do and identify that with someone with good desires, we are only looking at the desires that are relevant to the type of situation at hand, not all desires. There may be one, many or no people in existence, or who have ever existed, who would qualify as a good person in every type of situation to reasonably consider.

Secondly the use of "good" is entirely optional and, indeed, redundant. It is just a very useful shorthand - if it does not confuse. A good desire is a desire that tends to fulfil other desires. Still, why label that good? Well, surely it is in the interests of anyone and everyone who has a desire that could benefit from this desire, that is have or tend to have their desires fulfilled as a consequence of this desire, to promote this desire because it tends to fulfil their desires. This also applies to anyone and everyone who have others that they care or fear about who could also have or tend to have their desires fulfilled by this desire. That is most everyone have reasons to promote such a desire. Since most everyone, whether they have explicated this or not, thinks that fulfilling a desire is good, and that assistance in fulfilling their desire is also good, when one says that they have reason to promote such a desire, we could use a shorthand and say it is a good desire.

Does this resolve your confusion of good persons?

Right now that is out of the way lets look at the rest of your post.

DanielLC wrote:If my desire to play on the computer and my desire to stay well-fed are exactly two desires, then they are distinct. If they can be described as one desire: to play on the computer and be well-fed, they are not.

Desires can have more than one condition of fulfilment and it depends on the situation, but broadly painted as you have done, these are both normally distinct. Analysis of the specifics of situation usually make this clear. It is generally not at all fuzzy although success is not guaranteed but this is no different to any other area of valid empirical discourse.

What seems to be happening in our exchange, is that as I continue to lexically deal with your questions, you build further ones based on your own provisional answers with which I disagree or do not understand. I like to stop before we diverge too wildly.

So I will address one final point here:

DanielLC wrote:If you understand axiom four, ignore this. You never actually asked about it, but It's a new axiom, so I probably should explain it.
Axiom four essentially says that "a person results in something" is the same as "the act they'd take results in it". For example, if you, being in my situation, would result in, say, a certain donation to charity, that would mean that if I were to do what you'd do if you were in my situation, that charity would receive the same donation.

Yes I read that but for the reasons just stated never got to discuss this last time. It is not so much I did not understand it, but it does not really make grammatical sense, surely you could express this better and then we can see what relevance or use it has to your updated argument.

DanielLC · by **DanielLC** on 2009-01-09T00:05:00

I think part of the problem was me not explaining my questions well. Can you answer this one question for me? It was my main one for a while, but I suppose I didn't make that very clear.

Is whether a desire good or not at all dependent on the circumstances of the person with the desire? For example: If person A was allergic to penicillin and person B was not, and they were both sick, would the desire to take penicillin be a bad desire for person A but a good desire for person B, or would it just be a medium desire?

faithlessgod · by **faithlessgod** on 2009-01-09T09:17:00

DanielLC wrote:I think part of the problem was me not explaining my questions well. Can you answer this one question for me? It was my main one for a while, but I suppose I didn't make that very clear.

It is a good and very important question. Issues of clarity apply to both sides, I try my best but I am busy moving at the moment and replying between packing boxes.

DanielLC wrote:Is whether a desire good or not at all dependent on the circumstances of the person with the desire? For example: If person A was allergic to penicillin and person B was not, and they were both sick, would the desire to take penicillin be a bad desire for person A but a good desire for person B, or would it just be a medium desire?

First lets drop "good", we can add this back later. Remember this approach is reductive, as, I believe, are all utilitarian theories -well certainly objective ones?

Agent A has a desire to get better
A is not allergic to penicillin
A believes that taking penicillin is a means to fulfil this desire
Agent A takes penicillin and gets better, the desire is fulfilled

Agent B has a desire to get better
B is allergic to penicillin
B believes that taking penicillin is a means to fulfil this desire
Agent B takes penicillin and gets worse, the desire is thwarted

Everyone acts to fulfil their desires, given their beleifs.
If agent B did not know they were allergic to penicillin, then they were operating with a false belief, if the belief were corrected they would not take penicillin.

Now you are asking about the desire to take penicillin not the desire to get better. The desire to take penicillin as a means to fulfil the desire to get better. (The former is a desire-as-means, the latter is a desire-as-ends). In the case of A they have accurate beliefs so this would be a good desire - it is such as to fulfil the desire to get better. In the case of B they have inaccurate beliefs and this desire would not be such as to fulfil the desire to get better, that is it is a bad desire.

Does this help?

Does this help?

faithlessgod · by **faithlessgod** on 2009-01-09T09:22:00

DanielLC wrote:I think part of the problem was me not explaining my questions well. Can you answer this one question for me? It was my main one for a while, but I suppose I didn't make that very clear.

It is a good and very important question. Issues of clarity apply to both sides, I try my best but I am busy moving at the moment and replying between packing boxes.

DanielLC wrote:Is whether a desire good or not at all dependent on the circumstances of the person with the desire? For example: If person A was allergic to penicillin and person B was not, and they were both sick, would the desire to take penicillin be a bad desire for person A but a good desire for person B, or would it just be a medium desire?

First lets drop "good", we can add this back later. Remember this approach is reductive, as, I believe, are all utilitarian theories -well certainly objective ones?

Agent A has a desire to get better
A is not allergic to penicillin
A believes that taking penicillin is a means to fulfil this desire
Agent A takes penicillin and gets better, the desire is fulfilled

Agent B has a desire to get better
B is allergic to penicillin
B believes that taking penicillin is a means to fulfil this desire
Agent B takes penicillin and gets worse, the desire is thwarted

Everyone acts to fulfil their desires, given their beleifs.
If agent B did not know they were allergic to penicillin, then they were operating with a false belief, if the belief were corrected they would not take penicillin.

Now you are asking about the desire to take penicillin not the desire to get better. The desire to take penicillin as a means to fulfil the desire to get better. (The former is a desire-as-means, the latter is a desire-as-ends). In the case of A they have accurate beliefs so this would be a good desire - it is such as to fulfil the desire to get better. In the case of B they have inaccurate beliefs and this desire would not be such as to fulfil the desire to get better, that is it is a bad desire.

Does this help?

Does this help?

faithlessgod · by **faithlessgod** on 2009-01-09T11:02:00

A simpler (?) and complementary answer to my previous post.

DanielLC wrote:Is whether a desire good or not at all dependent on the circumstances of the person with the desire?

It depends on the desires. Good is a value and all values are relations between a desire and a state of affairs. Different desires refer to different types of states of affairs.

DanielLC wrote:For example: If person A was allergic to penicillin and person B was not, and they were both sick, would the desire to take penicillin be a bad desire for person A but a good desire for person B, or would it just be a medium desire?

In this case the states of affairs, that the desire refers to, includes the circumstances of the agent, that is the agent's circumstance is part of the states of affairs. This is because in evaluating any desire, states of affairs are not specified by desire under evaluation - but rather the states of affairs are specified by whatever (other) desires the evaluation is for. In this case this is the desire for health or an aversion to sickness and the state of affairs is the functional status of the agent. So, in the case of an agent who is allergic to penicillin, when one asks whether the desire to take penicillin is a good desire, this can be read as asking is this such as to fulfil the desire for the agent to be healthy or to regain health? Since the agent is allergic the answer is no. So it is a bad desire.

The idea of a "medium" desire is based on the mistake of averaging across cases for which averaging makes not sense, since these desires are not fungible, since they are agent-relative (there is no point in giving me penicillin if you need it). If we were to really on averages then we all should have less than two legs (since some people have lost their legs and so bring the average down below two).

faithlessgod · by **faithlessgod** on 2009-01-09T18:50:00

DanielLC wrote:Also, while you're at it, can you answer the last two questions on the main article?

"Why should you weaken the desire to torture the child, rather than weakening the child's aversion to torture?"
First lets us remove the demographic bias and say there is one sadist and one child with an aversion to torture

1.Sadist S has a desire to torture
2.Child C has an aversion to torture
3.Only C is available to S to fulfil S's desires
4.If S's desire if fulfilled, C's aversion is thwarted
5.If C' aversion is fulfilled, S's desire is thwarted
It looks balanced either way does it not? Which to prefer 5 or 4?

Here is one answer - there may be better ones:
There is a difference.
6. In 4 S is proactive and C is reactive.
7. In 5 C is not proactive - we do not know how this occurs but it is due to exogenous factors.
An aversion is an attitude to make or a state of affairs false, whereas a desire is to make or keep a state of affairs true. Of course you can convert one to another so an aversion to torture is a desire not to be tortured. So is a desire to torture also aversion not to torture? The negative qualifiers here reflects and helps to distinguish instigator versus the recipient. (This is really a conjecture I would welcome counter arguments to this, it sounds plausible to me).

This is one way to resolve an apparent parity dilemma as in 4 vs 5, the desire to be evaluated is the instigating desire, compare its existence with its absence, since it is this and only this desire that sets up the apparent dilemma in the first place.

I could probably have said this simpler but it would take more time to edit it down.

PhilosopherKing · by **PhilosopherKing** on 2009-12-19T05:59:00

The question was posed:
Why ought we weaken, or eliminate, the desire of 1000 sadists to torture an infant, as opposed to weakening, or eliminating, the infant's aversion to torture?

Suppose there are two men, X and Y, on a deserted island. X is struck with the insane urge to torture Y. Y has an aversion to such an act. Why would it be wrong to thwart X's desire? What would X say if you asked him, if the shoe were on the other foot, if he would be opposed to Y torturing him? Both X and Y share the same aversion to the other torturing them. X cannot universalize his action. Therefore, he is not justified in fulfilling his desire to torture Y. Both of their desires--their aversion to being tortured--are fulfilled in this way.

So, I am sure that each of the 1000 sadists desire a society that protects them from being tortured by the others, over the desire to torture an infant. None of them would prefer to be in the infants shoes.

I am new to Desire Utilitarianism, but is this a satisfactory answer?

DanielLC · by **DanielLC** on 2009-12-20T00:36:00

That seems rather silly. Suppose I wanted to take a girl on a date. I'd never want a guy to take me on a date. Does that make it a bad desire?

Similarly, supposing the infant had no aversion to torture in the first place, is the desire to torture it still bad?

This reminds me, if DeBeers advertises to make people want diamonds, and then sells them diamonds, is that good or bad?

Wow, looking back at this I really suck at speaking simply. I can barely understand this and I wrote it.

RyanCarey · by **RyanCarey** on 2009-12-20T02:37:00

That seems rather silly. Suppose I wanted to take a girl on a date. I'd never want a guy to take me on a date. Does that make it a bad desire?

Well in that instance the desire to take a person of the opposite sex on a date might count morally but the desire to take a female on a date would not, right? That's a big part of the idea of universalisability, I think. You can't build discrimination on the basis of race, gender, species, and so on into the rules or desires.

PhilosopherKing · by **PhilosopherKing** on 2009-12-21T04:18:00

"That seems rather silly. Suppose I wanted to take a girl on a date. I'd never want a guy to take me on a date. Does that make it a bad desire?

Similarly, supposing the infant had no aversion to torture in the first place, is the desire to torture it still bad?"

I can't speak for Alanzo Fyfe, and I am just starting to understand his theory. However, it is my understanding that, in Desire Utilitarianism, a desire is 'good' to the degree it fulfills, or tends to fulfill, other desires. A 'bad' desire is one that thwarts, or tends to thwart, other desires. In the case of X and Y, if X has the desire that "I am torturing Y" is true, and Y has the desire that "I am living relatively free of pain on a island with my friend X" is true, then X's desire would thwart Y's desire.

Now, does Alanzo Fyfe mean that desires are a good in and of themselves? or simply good as a means to the fulfillment of other desires? I really don't know at this point. I have to do some more reading and thinking. Hopefully, you and others can help me on this.

What I was suggesting in my post is that it is possible that the desire to torture the infant might be considered a bad desire because none of the sadists would desire the others to torture him. If it is established as a general principle that anyone, including infants, can be tortured on a whim, then a sadist may be tortured against his will.

True, you may want to take a girl out on a date. Furthermore, you would not want to date a man who wants to take you out. Your desire becomes a 'bad' desire if you take her out despite her aversion to it. You would be thwarting one of her desires. This is only my understanding. Does Alanzo Fyfe mean to imply desires are 'good' or 'bad' relatively, or absolutely? I don't know. It is my thinking that if you cannot universalize a desire, then it probably isn't a good desire. Maybe possible desires need to be considered, eg. If I were in such and such a circumstance, then what would I prefer?

faithlessgod · by **faithlessgod** on 2009-12-25T12:20:00

PhilosopherKing wrote:The question was posed:
Why ought we weaken, or eliminate, the desire of 1000 sadists to torture an infant, as opposed to weakening, or eliminating, the infant's aversion to torture?

Hi PhilospherKing

My ideas on this have evolved hopefuly both more concise and clearer. A few preliminary points:
I call this approach desrism rather than desire utilitarianism. I do this for two reasons:
a) It is easier to type!
b) This should reduce any confusions with classical act utilitarianism (AU), which is not

To expand on reason (b). Desirism is two steps beyond rule utilitarianism.
1. Desires are like rules except they cannot be broken (they can be over-ridden but only by other desires) so this does not collapse into AU
2. There is no utility to maximise, since utility is plural, incommensurate and indeterminate.

Suppose there are two men, X and Y, on a deserted island. X is struck with the insane urge to torture Y. Y has an aversion to such an act.

Good. Here you are capturing the idea of avoiding the distortions caused by the historical contingencies - leading to a certain distribution of ideas in a population - why should any distribution be better than any other is really the question of morality we are trying to answer. (I would drop the "insane" qualifier though as this is bringing in a prejudice unnecessarily into the analysis?)

Why would it be wrong to thwart X's desire? What would X say if you asked him, if the shoe were on the other foot, if he would be opposed to Y torturing him? Both X and Y share the same aversion to the other torturing them. X cannot universalize his action. Therefore, he is not justified in fulfilling his desire to torture Y. Both of their desires--their aversion to being tortured--are fulfilled in this way.

Yes in one sense we are looking (to promote) for universally desirable consequences over universally undesirable consequences. Your universalisation argument looks somewhat Kantian but that is not necessarily a bad thing.

The question here is to consider X's desire and compare it to its absence. Its absence leads to no desire thwarting, whereas it presence does. Now we could also ask the same of Y, however we can only modify using the social forces (morality) malleable desires and, usually, Y's desire is not malleable. Further Y having this desire does not pro-actively affect any other desires, it is a reactive desire whereas X's is necessarily a pro-active desire dependent on affecting other desires for its fulfilment. Finally if Y felt no pain then X's desire would not be fulfilled by acting on Y since part of its conditions of fulfilment are that it cause pain (I can expand on this last point if challenged ;-)

)

So, I am sure that each of the 1000 sadists desire a society that protects them from being tortured by the others, over the desire to torture an infant. None of them would prefer to be in the infants shoes.

Going from desire qua desire or X and Y on Island to 1000 sadists is going from individuals to groups and this is a different category (or level) of evaluation. The individual conclusion still applies but know we are applying it a group with a specific distribution of desires and are asking is this distribution of desire justified?

One problematic characteristic of social groups is institutionalised double standards and, in this case, it is just as likely that the sadists are or expect protection from torture whilst having no qualms about applying it to their socially acceptable victims. Notions of universality can vary (rationally unjustifiable but still) so that an in-group has benefits and protections that the out-group does not have or is denied. This is why I have issues with the "I am sure" argument above, does this make sense?

I am new to Desire Utilitarianism, but is this a satisfactory answer?

Welcome to experimenting and exploring these ideas. Does my response help?

PhilosopherKing · by **PhilosopherKing** on 2009-12-30T03:20:00

Faithlessgod,

I am a little confused about your post and was wondering if you could help me understand.

"1. Desires are like rules except they cannot be broken (they can be over-ridden but only by other desires) so this does not collapse into AU
2. There is no utility to maximise, since utility is plural, incommensurate and indeterminate."

I am unclear about the first point. How would one mistakenly assume DU collapse into AU? I am confused as to how one might think DU is AU. Also, on the second point, I thought that a good desire is one that maximizes the fulfilment of other desires. Is it the goal of the agent to maximize the number of desires to fulfill?

"The question here is to consider X's desire and compare it to its absence. Its absence leads to no desire thwarting, whereas it presence does. Now we could also ask the same of Y, however we can only modify using the social forces (morality) malleable desires and, usually, Y's desire is not malleable."

I understand you to be saying here that society can change X's desire, but not Y's. Therefore, we ought to change X's desire as opposed to Y's. But what if we could change Y's desire to live relatively free of pain?

Forgive me for being ignorant on these things. I haven't studied this form of U enough to have a good grasp on it. I just have a lot of questions. Like, is the fulfilment of a desire, or a desire, good-in-itself?

faithlessgod · by **faithlessgod** on 2009-12-30T10:13:00

Hi, some good questions

PhilosopherKing wrote:I am unclear about the first point. How would one mistakenly assume DU collapse into AU? I am confused as to how one might think DU is AU. Also, on the second point, I thought that a good desire is one that maximizes the fulfilment of other desires. Is it the goal of the agent to maximize the number of desires to fulfill?

The second half of this paragraph answers the first half!

DU cannot collapse into AU but DU is often misunderstood as DFAU - desire fulfilment act utilitarianism, an AU where the utility is desire fulfilment.(It is to minimise such misunderstandings that I renamed this" desirism")
DFAU: The right act is the act which maximises desire fulfilment
DU: The right act is the act which a person with good desires - desires that tend to fulfil other desires - would do

I understand you to be saying here that society can change X's desire, but not Y's. Therefore, we ought to change X's desire as opposed to Y's. But what if we could change Y's desire to live relatively free of pain?

There are many points here.
1)Addressing the specifics of torture, since part of X - the desire to inflict torture - is constituted by the inflicting of pain on others, if no pain is being inflicted then it is not torture!
2) In a sci-fi scenario, the creation of infants without the capacity to feel pain (that is an actual condition - a friend of mine has it) - is far beyond the employment of social forces to mould desires - using genetic modifications and/or neuropharmacology- so this is beyond the scope of morality as traditionally understood
3) Still granted (2) such infants (later adults) still have many other desires (as my friend does) that would be thwarted by being tortured, which would still make X a bad desire.
4) In order to create an infant without any desires that could be thwarted by torture, I would argue that you have created an organism without any desires at all, one that is neither a moral agent nor capable of being evaluated as moral recipient (as one can do for animals who are not moral agents). Again part of what is constituted by torture is the thwarting of the desires of others for one's own benefit, so again this is not torture!

However I think all these points above, implied by some aspects of yours (and others) previous posts are side issues. Another way of formulating the core question is to ask whether someone with good desires would act so as to promote or inhibit the desire to torture. Since promoting this desire tends to lead to the thwarting of desires and inhibiting it - creating and encouraging an aversion to torture - tends to lead to the reduction of thwarting of desires, the answer is they would seek to inhibit the desire to torture, hence the desire to torture is a morally bad desire and an aversion to torture is a morally good desire (this does not apply to psychologically healthy consenting adults of course e.g sado-masochism).

There is more I could add but lets stop here for now. Hope this helps, back to you.

PhilosopherKing · by **PhilosopherKing** on 2010-01-01T09:25:00

"DFAU: The right act is the act which maximises desire fulfilment
DU: The right act is the act which a person with good desires - desires that tend to fulfil other desires - would do"

I am still a little unclear as to the difference between these two schools of thought. Could you please give an example?

RyanCarey · by **RyanCarey** on 2010-01-01T14:01:00

I think acts are only judged indirectly, that is in terms of their desires. From what I gather, DU arises from an odd metaphysical premise that ethics can only shape desires, not acts. Hence, DU is framed entirely in terms of desires. A good desire is one that fulfils other desires.

faithlessgod · by **faithlessgod** on 2010-01-01T17:14:00

PhilosopherKing wrote:I am still a little unclear as to the difference between these two schools of thought. Could you please give an example?

It is a question of the primary evaluation focus. AU directly evaluates acts and everything else indirectly in relation to (individual instances of) acts. DU directly evaluates desires and everything else indirectly in relation to desires (which leads to a variety of different acts in different circumstances). In addition to DFAU being a form of AU where the utility is DF, by contrast DU has no utility to maximise.

faithlessgod · by **faithlessgod** on 2010-01-01T17:25:00

RyanCarey wrote:I think acts are only judged indirectly, that is in terms of their desires... A good desire is one that fulfils other desires.

Correct

From what I gather, DU arises from an odd metaphysical premise that ethics can only shape desires, not acts. Hence, DU is framed entirely in terms of desires

Interesting. DU arises from what is neither an odd nor metaphysical nor a premise "that ethics can only shape desires". Rather it is based on the empirical evidence - from both philosophical and cognitive psychology and from social evidences -as to how the processes of praise and condemnation could work. It is based on the most standard and accepted (so least odd) and emphatically non-metaphysical and pragmatic conclusions (not premise) in these fields. Further it is not that "ethics can only shape desires" but that the social forces (praise, condemnation etc.) can only directly shape desires (and only indirectly acts and so on) .

DanielLC · by **DanielLC** on 2010-01-01T18:32:00

Each individual person has their own utility, which they attempt to maximize. This includes someone with good desires. As such, DU has utility. It's the utility that someone with good desires would have. Their utility would be equal to others' DF, which means that DU is AU in which utility is DF. In other words, it's DFAU.

Did I mess up somewhere?

faithlessgod · by **faithlessgod** on 2010-01-03T14:20:00

DanielLC wrote:Each individual person has their own utility, which they attempt to maximize. This includes someone with good desires. As such, DU has utility. It's the utility that someone with good desires would have. Their utility would be equal to others' DF, which means that DU is AU in which utility is DF. In other words, it's DFAU.

Did I mess up somewhere?

An assertion or assumption that DU just is DFAU is an illegitimate argument, one I all too often have to deal with. However here you are trying to make an argument that DU is DFAU, this is quite legitimate. However I do not think it works for the following reasons.

1. DU does not deny utility, only that it accepts Mackie's argument from the indeterminacy of utility, which makes it impossible to maximise. You are implicitly accepting this point when you say everyone has their own utility. These can be incommensurate, incompatible and plural. Indeed utility is a type of value, and the desire fulfilment theory of value supports/agrees with Mackie's argument.

2. If we temporarily grant your notion of utility, it still does not follow that every good person has the same utility to maximise. They can still all have different utilities. DU does not need to make an additional assumption that these are the same. Indeed it is cognizant of the psychological facts which denies empirical support to such an additional assumption.

3. Regardless, what "good people" have in common is a sensitivity to the environment, which is partly constituted by morality - the social forces - and they go about realising their "individual utilities" without adversely contributing to this environment.

4. DF is orthogonal to your (or anyone's) conception of utility. It is a property that relates brain states to states of the world - the relation being "fulfilled", "thwarted", "in progress", "not started" etc. Anyone's "utility" is about those valued end states of the world, however morality is focused on (dis)valueable means to bring them about.

DanielLC · by **DanielLC** on 2010-01-03T22:16:00

1. You say a good desire is one that fulfills other desires. If desire fulfillment is incommensurate, and you have a desire that fulfills some peoples' desires but not others', how can you tell if it's a good desire? For example, if you live in a house with two other people, one who wants to paint the house red and one who wants to paint it blue, would wanting to paint it blue be good, bad, or neutral? What if there's one person who wants it red and six billion people who want it blue?

2. I immediately assumed based on good/bad desires that there was some best desire that you really ought to be doing. It looks like you're saying that this was an incorrect assumption, and DU only gives you several possible things to do, rather than one best thing you should do the most. Is this correct?

4. Although DF is entirely about the means, it seems to define the means based on how well they accomplish certain ends (fulfilling others' desires), making it effectively about those ends.

faithlessgod · by **faithlessgod** on 2010-01-04T01:55:00

DanielLC wrote:1. .... If desire fulfillment is incommensurate,...

(Desire) fulfilment is the name of the status of a relation between brain states and states of the world, "incommensurate" does not apply. What can be incommensurate are the targets (states of the world) of different peoples desires (or conditions of fulfilment).

2. I immediately assumed based on good/bad desires that there was some best desire that you really ought to be doing. It looks like you're saying that this was an incorrect assumption, and DU only gives you several possible things to do, rather than one best thing you should do the most. Is this correct?

People seek to fulfil the more and stroonger of their desires -whatever they are. The question is how they act to fulfil them, that is the issue. So the idea of a best desire does not apply. This is about means not ends.

4. Although DF is entirely about the means, it seems to define the means based on how well they accomplish certain ends (fulfilling others' desires), making it effectively about those ends.

It is the other way, it is about turning ends into means. Given anyone's ends , these can be evaluated by turning them into means, when the question is a moral one, the scope is the effect on everyone's ends.

Does this help?

PhilosopherKing · by **PhilosopherKing** on 2010-01-05T05:51:00

Faithlessgod
I have some more questions:

How does DU define a good agent?
What is a good desire according to DU?

Thank you for your patience. It takes me a while to understand some concepts.

PhilosopherKing · by **PhilosopherKing** on 2010-01-05T06:10:00

Also, what is the role of pleasure in DU?

faithlessgod · by **faithlessgod** on 2010-01-05T20:09:00

PhilosopherKing wrote:Faithlessgod
I have some more questions:

How does DU define a good agent?

A good person is a person with good desires

What is a good desire according to DU?

A desire is morally good to the degree it tends to fulfil and not thwart other desires

Of course the term "good" is optional, those desires exist regardless.

faithlessgod · by **faithlessgod** on 2010-01-05T20:14:00

PhilosopherKing wrote:Also, what is the role of pleasure in DU?

Pleasure can be a target of a desire, the desire is fulfilled when the states of affairs is true, when the agent -presuming it is his or hers' pleasure that was being sought - is experiencing pleasure. However there are many possible targets of desire, and many ways of achieving pleasure, as to whether the seeking of a certain pleasure is morally justified then one again evaluates that desire by its material effect on other desires.

Preference Satisfaction versus Desire Fulfilment

Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment

Re: Preference Satisfaction versus Desire Fulfilment